odeling the "ether," a fluid medium through which forces were alleged to propagate, was a major preoccupation for British physicists throughout the nineteenth century. One of the earliest physicists to posit an ether as the propagating medium for electromagnetic forces was Michael Faraday, who introduced the concept of "lines of force" to demonstrate how electric and magnetic forces are transmitted between particles in an ambient medium, and to represent the disposition of these forces in space. William Thomson elaborated upon Faraday's work by arguing that the electromagnetic field could be represented as a system of vortices in the ether; later, he viewed the ether as a plenum in which forces acted in an ethereal continuum. James Clerk Maxwell translated Thomson's vortices into a mechanical model that represented the action of the field in transmitting forces by the action of particles in the ether.
However, whereas Thomson developed a theory of an ethereal continuum as a physical embodiment of Faraday's plenum of force, Maxwell explored the geometrical implications of Faraday's lines of force. Maxwell's interpretation of Faraday presupposed absolute space that was not determined by the disposition of the field; the field, represented geometrically by lines of force, was in space, and space was a fundamental ontological category. A physical representation of the field therefore required an ambient ether as the substratum (but not the constituent) of the field.
Finally, Maxwell intended his models as an illustration, not an ultimate physical explanation. In the end, he abandoned models altogether. While continuing to emphasize that electromagnetic phenomena were produced by the motions of particles in a mechanical ether, he used methods of Lagrangian analytical dynamics to represent the electromagnetic field instead. His final formulation of the field in the Treatiseconcluded that mechanical modeling was too imprecise to be useful: "The problem of determining the mechanism required to establish a given species of connexion between the motions of the parts of a system always admits of an infinite number of solutions." Others, such as Fitzgerald, wished to jettison the idea of material ether altogether, or at least limit the claims an ether model could make when describing the microstructure of matter. Fitzgerald's intended his model of the ether, composed of rubber bands and wheels to show that it was possible to represent the connection between ether and matter by means of a mechanism, thereby providing a mechanical illustration of ether strain (when matter was in motion) — but not an explanation of physical reality.
Thomson and Tait
William Thomson's and P. G. Tait's Treatise on Natural Philosophy (1873) was an important source for Maxwell's Treatise. Lagrangian mathematical methods allowed Thomson and Tait to avoid reference to concepts of momentum, velocity, and energy after they had been replaced by the appropriate symbols in generalized equations of motion. Maxwell's work was in part an attempt to re-interpret Thomson and Tait's Treatise in terms of dynamics, in which the field was a structure of matter in motion, and action was communicated between contiguous parts of the system. Because the field was a moving system, changes in its structure and motion would lead to the communication of action — would lead, in short, to an observable effect. It would be possible, thereby, to avoid postulating unobservable forces between the micro-structures of the field, as well as to avoid the definition and representation of these micro-elements. The inertia, density, and mobility of an incompressible universal plenum were sufficient to define the properties of the vortex atom (pace Thomson) without further arbitrary assumptions.
The decline of mechanism
ne unexpected implication of Maxwell's work was that the velocity of electromagnetic waves propagated in the ether was found to be identical to the velocity of light, leading Maxwell to conclude that electromagnetic and luminiferous ethers were the same. In this electromagnetic theory of light, light was considered an electromagnetic vibration in the ether. In order for this theory to be correct, however, ether had to be rare enough not to impede planetary motion, and rigid enough to support the undulations of which most physicists, by 1830, thought light to consist.
The complications did not stop there. Whereas the wave theory of light and electromagnetism pointed to an elastic ether subject to distortion, the Kerr magneto-optic effect (the production of a slight elliptic polarization, produced when plane-polarized light is reflected from the polished pole face of an electromagnet) indicated that the ether was rotational, a property most satisfactorily represented by vortices in a fluid. To explain this effect, the Maxwellian J. Larmor posited a physical model of a fluid, rotationally elastic, and incompressible ether, but this model too was meant merely as a working representation of how the ether might work, rather than as a model of the microstructure of matter. Larmor saw no connection between ether and matter. For Larmor, all that could be known about the ether could be formulated as differential equations defining the properties of a continuous medium; because ether was prior to matter, it was not expressible in material terms. The fundamental unit of matter was a center of strain in the vortex ether; Larmor supposed these centers were endowed with an electric charge and called them electrons.
German physicist Heinrich Hertz's experimental detection of electromagnetic waves in 1888 was regarded as definitive experimental confirmation of the electromagnetic field. Hertz advocated taking Maxwell's equations as prior, and changing models as necessary, in order to bring the models into conformity with Maxwell's equations. By the end of the century, and largely as a result of the sweeping Maxwellian synthesis, electrodynamics had replaced mechanics in providing the conceptual foundations of physics. In this new electrodynamic ontology, the field had an independent physical reality denuded of mechanical properties.
Last modified 2 April 2002